Ellen Theisen Executive Director John Gideon Information Manager

www.VotersUnite.org

September 13, 2004

Nick Handy Director of Elections 1007 S. Street Olympia, WA 98504-0237

Dear Mr. Handy,

Quite by accident, I learned that you had written a response to my letter of August 23 and may have distributed it to the 200 State leaders I copied on my letter. Fortunately, Representative Jim Buck's office was kind enough to fax me their copy of your response. I have not yet received a copy from you. Since your letter does not indicate the list of recipients, I am sending my reply to all the leaders I originally contacted. I am also mailing you a copy by U.S. Post.

I do not understand why you are willing to expend so much effort defending decisions made by Secretary of State Reed, yet you are not willing to take the obvious steps that experts have stated are necessary to ensure the integrity of our 2004 primary and general elections — conducting a robust manual audit of the actual ballots cast in the primary election and replacing the unexamined software with the fully certified software for the November election.

You say that you "are not aware of any evidence suggesting that elections in Washington are at risk." While there may not be evidence of error or fraud — yet — let me clarify for you the evidence of risk I presented in my previous letter:

1) The names and parties of most of the candidates were public knowledge before software development began, thus increasing the ease with which malicious code could be inserted into the software during the development process to pervert the outcome. Since no independent tester has had an opportunity to examine the software source code, such malicious code would be impossible to detect.

2) The software was developed rapidly and the vendor's testing was limited. Both of these conditions significantly increase the risk of error. In fact, the Statement of Work from ES&S to Pierce County cited both of these concerns as the reasons they were unwilling to be liable for defects found in the software.

3) The provisional certification process adopted by Secretary Reed for this fall's elections omitted the independent testing that the Secretary has required for all previous certifications. He has repeatedly claimed the votes of Washington State's voters are safeguarded precisely because of that independent testing. If the independent testing was necessary for previous certifications, why was it not necessary for these recent ones? If it was not necessary before, why did he require it? Your response to my concerns fails to address the significant financial and legal risk the provisional certification poses to Washington State. Since Secretary Reed's installation of the software with a high risk of error and fraud violated his own July 7, 2004 "Policy on Electronic Voting Systems," as well as the testimony of both Secretary Reed and David Elliott, the previous State Elections Director, before committees of the House and Senate in January of this year, his actions leave the state extremely vulnerable to valid legal challenges by candidates after the election.

So, beyond even the risk of election error and fraud, the Secretary's actions have created a situation in which all of Washington's voters could be disenfranchised for days, weeks, or months. Even worse, in the case of the Presidential Election, the votes of Washington’s citizens could be ignored permanently if the delegation to the Electoral College cannot be certified in time.

In your response, you cite Dr. Douglas Jones as an expert who recognizes that there are rare circumstances when it is necessary to install a patch without going through the national certification. Let me remind you of the rest of the statement Dr. Jones sent to me regarding the recent changes in Washington state's election systems:

This is not a good procedure. No software upgrade should be allowed without going through the ITA process. In the rare event that circumstances require late patching of a voting system, and particularly if certification is waived or done on a rush basis, additional defenses such as California-style random recounts (for paper ballots) or parallel testing (for direct-recording voting systems) should be required.

Note that Dr. Jones is discussing circumstances that require late patching. You cite the use of the consolidated ballot as the reason for the installation of the patch, but the consolidated ballot was a choice made by the six counties, not a requirement of the new primary law. Secretary Reed chose to allow the use of the consolidated ballot and chose to waive the state's long standing certification standards in order to accommodate its use.

So, in this case, certification was both waived and done on a rush basis. Note also that California's audit requirement for fully certified software is over 1%, that is, approximately 1000 times more than the 0.0015% audit the Secretary is encouraging the political parties in King County to request as a check on the accuracy of the provisionally certified software recently installed.

A report from a recent Harvard University symposium on voting systems supports outcome audits, like those recommended by Dr. Jones, as a standard operating procedure, even when robust testing is conducted and equipment is fully certified:

Testing is necessary but not sufficient for a well-run election. Testing is never perfect, as it can overlook certain factors or interactions that may be easier to detect in hindsight. Systems interact with each other in unpredictable ways, often impossible to detect in a reasonable battery of tests.1

The report unequivocally states, "testing before the vote cannot verify accuracy of final tally," and declares, "Equipment testing does not displace the need for outcome auditing."

1 Voting, Vote Capture & Vote Counting Symposium. June 2004. Kennedy School of Government Harvard University. Electronic Voting Best Practices. A Summary. p. 22. http://designforvalues.org/voting/ABPractices.pdf

2 In addition, a report recently commissioned by King County and prepared by the King County Citizens' Election Oversight Committee recommends manual audits even under normal circumstances to improve the standard procedures and increase public confidence. The report states:2

No machines or procedures are totally foolproof, and it is a good idea to double check. Questions of a system’s accuracy can be verified with regular audits, and regular “human” audits become more necessary as technological complexity increases.

Your primary duty is to carry out fair and accurate elections in order to safeguard citizens votes. When you are presented with evidence that the votes may not be secure, you are obligated to undertake measures to correct the problem.

On behalf of the growing coalition of Washington State organizations I represent, I insist on a robust 16% manual audit of the actual ballots cast in the primary on the provisionally certified software, even if this means convincing Secretary Reed to establish an emergency Washington Administrative Code to allow the audit.

Using the provisionally certified patch in the general election is unjustifiable as it puts our votes and our State at risk. On behalf of those same organizations and to protect the franchise of the voters of Washington State, I also insist that the fully certified software be restored for use in the general election in November.

Sincerely,

Ellen Theisen (360-437-9922) John Gideon (360-377-4925)

CC: Governor Gary Locke Washington State Senators and Representatives Washington State's Congressional Delegation King County Executive Ron Sims Secretary of State Sam Reed County Council Members of all six counties Attorney General Washington Democratic Party Chairman Paul Berendt King County Director of Elections Dean Logan Washington Republican Party Chairman Chris Vance Thomas T. Osinski Jr., Attorney at Law Dr. David Dill Dr. Douglas Jones Dr. Barbara Simons

2 Report on King County Elections.. May 24, 2004. King County Citizens’ Election Oversight Committee. p. 52. http://www.metrokc.gov/mkcc/docs/CEOC_report.pdf

3